Deriving initial data warehouse structures from the conceptual data models of the underlying operational information systems

Abstract

In recent years the construction of large scale data schemes for operational systems has been the major problem of conceptual data modeling for business needs. Multidimensional data structures used for decision support applications in data warehouses have rather different requirements to data modeling techniques. In case of operational systems the data models are created from application specific requirements. The data models in data warehouses base on the analytical requirements of the users. Furthermore, the development of data warehouse structures implicates the consideration of user-defined information requirements as well as the underlying operational source systems. In this paper we show that the conceptual data models of the underlying operational information systems can support the construction of multidimensional structures. We would like to point out that the special features of the Structured Entity Relationship Model (SERM) are not only useful for the development of big operational systems but can also help with the derivation of data warehouse structures. The SERM is an extension of the conventional Entity Relationship Model (ERM) and the conceptual basis of the data modeling technique used by the SAP Corporation. To illustrate the usefulness of this approach we explain the derivation of the warehouse structures from the conceptual data model of a flight reservation system.